A total of 141,260 domestic sheep (Ovis aries; 98,638 male and 42,622 female) and 37,399 domestic goats (Capra hircus; 32,441 male and 4,958 female) were slaughtered and investigated between September 2021 and March 2022 in 6 regional abattoirs (Sulaimany, Piramagroon, Khalakan, Darbandixan, Takya and Soran) in the northern Iraq Kurdistan region (Supplementary material S1). All slaughtered animals were carefully inspected by specialist personal and in case of any pathological changes reinspected by veterinarians. If Sarcocystis macrocysts were detected in esophagus, diaphragm or skeletal muscles, all obvious macrocysts and the surrounding muscle tissue were collected, and age (as estimated by dental examination) and sex of the respective animal were recorded. All tissue sampling was performed under the supervision and with permission of the state veterinarian in the above-mentioned slaughterhouses. Written consent from the animal’s owners for the participation of their animals in this study was not required in accordance with the national legislation and the institutional requirements.

Out of a total of 141,260 sheep, only 20 had Sarcocystis macrocysts, and out of a total of 37,399 goats only nine carried Sarcocystis macrocysts, all of which were embedded in the muscular tissue and were milky-white and grainy in appearance. All 29 infected animals appeared healthy and showed no signs of disease before slaughter. Of the 20 infected sheep, we collected 21 pieces of infected tissue: 11 from skeletal muscle, five from the diaphragm, and five from the esophagus, namely one sheep had macrocysts in both skeletal muscle and esophagus. Of the nine infected goats, 12 tissue pieces were collected, nine from the esophagus and three from skeletal muscle. Here, three goats exhibited macrocysts in both the esophagus and skeletal muscle. Each tissue piece contained a minimum of two and a maximum of 12 Sarcocystis macrocysts. The size of Sarcocystis macrocysts in sheep and goats were then measured using a digital caliper. Tissue pieces from esophagus and skeletal muscle containing macrocysts were randomly divided into three groups, and then immediately fixated by immersion in the following solutions: 4% paraformaldehyde (PFA) in 0.1 M phosphate butter pH 7.4 for histological examination, 70% ethanol for DNA purification, and sequencing or 2% glutaraldehyde (GLA) plus 2% PFA in 0.1 M phosphate butter pH 7.4 for transmission electron microscopy.

Ten of the 21 macrocysts-containing sheep samples, and four of the 12 macrocysts-containing goat samples were fixed in 4% buffered formaldehyde, embedded in paraffin, serially sectioned at 5 μm thickness with a rotary microtome (Leica Autocut 1,140). The slides containing tissue sections were then stained with Tri-PAS (30 min 1% Periodic acid at room temperature 10 min Schiffś reagent (19) Mayer’s hematoxylin and methylene blue. All samples were further imaged with a Zeiss AX10 light microscope and a Hitachi HV-F202 camera. The following measurements were all measured using image analysis software Image-Pro® Plus version 6.0 (Media Cybernetics, Inc. MD 20910, United States): (i) thickness of the degenerating layer, primary and secondary cyst walls, villous protrusions, and internal cyst compartments, and (ii) bradyzoites dimensions.

Twelve Sarcocystis macrocysts from sheep and four from goats were collected from the esophagus and skeletal muscles and immersion-fixed at 4°C in a solution containing 2% GLA plus 2% PFA in 0.1 M Phosphate buffer, pH 7.4, and left in solution for up to 2 weeks for logistical reasons. After nine rinses in 0.1 M cacodylate buffer pH 7.4, the cysts were postfixed for 2 h with 1% OsO4 + 1.5% K3Fe³ (CN) 6 in 0.1 M cacodylate buffer, pH 7.4 (20, 21). The tissues were dehydrated via acetone/propylene oxide and embedded in epoxide resin (epon; Fluka, Switzerland).

For morphometric light microscopical analysis, 0.5 μm thick cross sections of epoxide embedded cysts were stained with methylene blue. Ultrathin sections (30 nm) were cut with a diamond knife, mounted on formvar/carbon-coated 200-mesh copper grids and contrasted with uranyl acetate and lead citrate solution. TEM was performed with a Zeiss EM109 (80 kV, 200 μm condenser and 30 μm objective apertures, TRS-2 K-Camera). Magnification of micrographs was calibrated by means of a cross-grating replica (2,160 lines/mm; Polaron, England).

To analyze morphology, methylene blue stained semithin sections of several macrocysts were imaged with a Zeiss AX10 light microscope. For the investigation of cyst wall morphology, villous protrusions, and bradyzoites, the length and width of these Sarcocystis components were measured using the same image analysis software as for paraffin sections (see above).

The collected macrocysts from sheep (n = 10) and goats (n = 4) were stored in 70% ethanol until DNA extraction. The samples were briefly lysed in lysis buffer (200 mM Tris pH8, 50 mM EDTA pH8; 100 mM NaCl, 1% SDS in H2O), supplemented with 30 U/mg Proteinase K at 600 rpm and 55°C overnight. The following day, samples were centrifuged 8 min at 14000 rpm to separate the genomic DNA in the supernatant from cell debris in the pellet. The supernatant was transferred to fresh tubes containing 500 μL ice cold 100% EtOH resulting in DNA filaments to form. For washing, the DNA filaments were collected with a Pasteur pipette tip from the liquid and transferred into a new tube containing 375 μL ice cold 70% EtOH. Finally, DNA was resuspended in 75 μL ultra pure nuclease-free water overnight at 42°C and stored at 4°C until further analysis.

New primers for amplification of the 18S rRNA and 28S rRNA genes were designed, so that one primer pair each could be used to amplify the gene of interest from all of the mentioned species (S. gigantea, S. medusiformis and S. caprafelis moulei). Therefore, the reference sequences for either the 18S rRNA or 28S rRNA genes of S. gigantea, S. medusiformis and S. caprafelis were aligned. Primer binding sites were selected in sections were all three reference sequences have 100% sequence identity and surrounding sections where the sequences had the greatest differences between the three named species. The primer pair oBM001 (5′- ACTGCGAATGGCTC ATTAAAACA-3′), and oBM002 (5’TGATCGTCTTCGAGCCCCTA- 3′) amplified a fragment between 1,001 bp and 1,072 bp from the 18S rRNA gene of each species. For the 28S rRNA, the primers oBM003 (5′-AGCGGTGGAGAAGAAAATAACA-3′) and oBM004 (5′-TCACATGGAACCCTTCTCCA-3′) resulted in a fragment of 1,674 bp, 1,676 bp, and 1,680 bp, respectively.

PCR amplification was performed in a thermocycler using the following program for 18S rRNA: Initial denaturation at 95°C (3 min), followed by 30 cycles of 95°C (30s), 60°C (30s) and 72°C (70s), ending with a final elongation at 72°C (5 min). For the 28S rRNA, the program differed slightly: Initial denaturation at 95°C (3 min), followed by 30 cycles of 95°C (30s), 58°C (30s) 72°C (1 m and 45 s), ending with a final elongation at 72°C (5 min).

PCR products were run on a 1% agarose gel together with GeneRuler 100 bp and 1 kb plus DNA ladders (Thermofischer), stained with 0,1 μg/mL Ethidiumbromide, and further analyzed after electrophoresis with a UV transilluminator. Proportionally correct bands were excised from the gel and underwent DNA extraction using the NucleoSpin Gel and PCR Clean-up kit (Macherey-Nagel), according to the manufacturer’s instructions.

For each Cyst, one PCR product per gene of interest was sequenced with forward and reverse primers for the respective gene in two individual reactions.

Sequencing results were obtained from Microsynth Seqlab ¹ and analyzed in SnapGene and with the NCBI Blast Tool. ²

Nucleotide sequences referred to as reference sequences or BLAST results mentioned in this paper are available in the GenBank™ NCBI nucleotide databases under the accession numbers: MK420020.1, MK420025.1, MK420021.1, MK420026.1, L76473.1, AF012884.1, KP053891.1, MK045326.1, and GQ131808.1. All our data sequences for 18S rRNA and 28S rRNA genes are deposited in NCBI under the accession numbers (see Supplementary material S1).

All values (dimentions of macrocysts, bradyzoites and villar protrusions, thickness of cyst walls and degenerating layers) are expressed as the mean ± s.e.m. (standard error of the mean). Following outlier analysis, two group comparisons were performed using an unpaired two-tailed t-test for normal distributed data and a Mann–Whitney test for nonparametric data. Corresponding details are given in the figure legend. Analysis were performed using GraphPad Prism 9.2.0. Statistical significance was considered at p < 0.05 and marked with *, p < 0.01 was assigned with **, p < 0.001 with ***, and p < 0.0001 with ****.

Out of a total of 141,260 sheep, and 37,399 goats (Supplementary material S1) slaughtered and subsequently examined by veterinarians, only 20 sheep (0.01%) and nine goats (0.02%) contained Sarcocystis macrocysts. S. gigantea and S. caprafelis moulei fresh, unfixed macrocysts appeared milky-white and grainy and were always found embedded in muscle tissue. S. gigantea macrocysts (Figure 1) appeared large, thick, round or elongated. They were located exclusively in the diaphragm and esophagus with an average length of 6.7 mm and width of 2.9 mm. Further, S. caprafelis moulei macrocysts (Figure 1D; Supplementary material S2) seemed also large, thick, oval or round; they were exclusively found in the esophagus–preferentially in the lower half–with an average length of 6.3 mm and width of 4.0 mm (Figures 2A,B; Supplementary material S2). A two-tailed t-test revealed no significant difference (p<0.05) in macrocyst length, but a significant difference (p<0.01) in macrocyst width (Figures 2A,B) between sheep and goats. This mean difference of 2.9 mm versus 4.0 mm macrocyst width is not, however, reliably discernible by observation. The macroscopic identification of S. gigantea and S. caprafelis moulei depends solely on the source of the sample: S. gigantea is collected from sheep, S. caprafelis moulei from goats. Evidence that these truly are different macrocyst subspecies is provided only by DNA sequencing and ultrastructural analysis (see below). All structures, that were deemed macrocysts by macroscopical inspection, proved to be so by histological, electron microscopical or DNA analysis.

In histological sections of muscle tissues, the macrocysts were thick-walled and consisted of two layers in both species. The mean wall thickness was 18.9 μm in S. gigantea and was 15.3 μm in S. caprafelis moulei (Figure 2E). An outer PAS-positive, i.e., covalently bound glycoconjugates containing secondary thick wall (11.6 μmin sheep; 8.0 μm in goats) was stained purple-red in Figures 1B,E, 2D. And an inner PAS-negative primary, relatively thick wall (7.3 μm in sheep and goats) was stained pale yellow in Figures 1B,E, 2C. Regarding the PAS-positive outer secondary wall, it is noteworthy that Apicomplexan protozoan parasites trigger host innate immune responses via their cell surface glycocojugates (22). On methylene blue staining, the secondary cyst wall appeared thick and bluish-pale in both species, whereas the primary wall was obviously thinner and purple in color. In both species, a vacuolated layer of degenerating host cells was found between the primary and secondary cyst walls (Figures 1C,F). The inner surface of the macrocysts within the primary wall was divided into a series of large round, oval or elongated compartments. The compartment sizes were approximately 1,065 μm² in S. gigantea and 917 μm² in S. caprafelis moulei (Figure 2F). Statistical analysis revealed no significant difference (p < 0.05) in the thickness of macrocyst walls and sizes of compartment between sheep and goats (Figures 2C,F). Compartments were packed with banana-shaped bradyzoites with an average size of 12.3 × 2.6 μm in S. gigantea (Figure 1C), and 13.9 μm´ 4.4 μm in S. caprafelis moulei (Figure 1F). Statistically, there is a significant difference (p < 0.05) in the length of bradyzoites and a highly significant difference (p < 0.0001) in the width of bradyzoites (Figures 2G,H) between sheep and goats. As with the macroscopical observations the differences in microscopical measurements do not allow the identification of S. gigantea versus S. caprafelis moulei, but only the source of the sample. Notably both S. gigantea from sheep and S. caprafelis moulei from goats have the same PAS-positive secondary outer cyst wall that is described here for the first time.

The S. gigantea and S. caprafelis moulei outer secondary cyst wall (PAS-positive in lightmicroscopy) consisted of a thick, fibrillar, collagenous layer, whereas the inner primary cyst wall had a homogeneous appearance. Between the secondary and primary cyst walls, was a layer of the degenerated myofibril remnants consisting of numerous vacuoles of various host cell remnants. The average thickness of the degenerated layer was 4.5 μm in S. gigantea (n = 40; Figures 3A,B) and 4.9 μm in S. caprafelis moulei (n = 38; Figures 3E,F). Statistically, no significat difference (p < 0.05) was observed between both species (Figure 2K). In addition, the primary macrocyst wall formed cauliflower-, trapezoid- or mushroom resembling protrusions with an average width of 2.4 μm, and a height of 3.1 μm (n = 27, S. gigantea; Figure 3B). S. caprafelis moulei also had similar protrusion features but was slightly larger with an average width of 3.5 and height of 3.4 (n = 25; Figures 2I,J, 3F). Statistical analysis revealed a high significant difference (p < 0.001) in the width of villar protrusions and a significant difference (p < 0.05) in their hight between S. gigantea and S. caprafelis moulei (Figures 2I,J).

In both species, numerous dendritic-resembling filaments extended from the primary cyst-wall surface and from available surfaces of some protrusions. These filaments were significantly longer and more numerous in goats than in sheep: (Figure 3C in sheep and Figure 3G in goats)–this is the only clear morphological difference between S. gigantea and S. caprafelis moulei. Moreover, within the cauliflower-resembling protrusions, long microtubules extended longitudinally transversely from the core of the free protrusions into the ground substance of the protrusions (S. gigantea: Figures 3B,C; S. caprafelis moulei: Figures 3F,G). Coarse, spherical, electron-dense granules occurred at the base and margin of these projections and in the ground substance of the protrusions (Figures 3C,G; S. gigantea and S. caprafelis moulei, respectively). Furthermore, no marked differences in the ultrastructure of the bradyzoites were detected between these two closely related sister species.

In this study, the fine structure of S. gigantea cyst wall (Figures 3A–C) resembled that of the same species described previously [Figure 8.2 (C) in (1)], and the cyst wall of S. caprafelis moulei resembled that described briefly in previous studies [compare Figures 1–2 in (17)] with our (Figures 3E–G). In both species, the ground substance had a homogenous appearance: it formed a 3–4 μm thick septa boarder, enclosing the bradyzoites. Immediately below the primary cyst wall, some bradyzoites were visible in the ground substance. Banana-shaped bradyzoites contained a bilayered pellicle, slightly thickened at the anterior end, with a large number of small, rod-shaped micronemes” tiny, apical (23), cigar-shaped organelles involved in host cell recognition and adhesion„ (23). At the anterior end of the bradyzoites, was a rounded end called a “conoid.” This is an intricate structure of spirally arranged microtubules in the apical complex of Apicomplexa that plays a role in host cell entry (cf. 23). Behind the micronemes, were approximately 8–12 irregular, varying rhoptries, along with a small amount of elongated, tubular mitochondria and amylopectin granules. Rhoptries, micronemes and electron-dense granules have previously been described as specialized secretory organelles from Apicomplexa parasites (23). The bradyzoites nucleus is mainly located at the posterior end of the body (Figure 3D in sheep, and Figure 3H in goats).